Skip to content

Latest commit

 

History

History
81 lines (59 loc) · 3.16 KB

README.md

File metadata and controls

81 lines (59 loc) · 3.16 KB

Linorobot2 Autonomous Navigation

This project leverages the Linorobot2 platform and the Nav2 framework to implement autonomous navigation and affective mapping in dynamic, unstructured environments. Our goal is to enable a simple four-wheeled robot to intelligently explore hazardous areas, build a map using SLAM, and navigate safely without human intervention. By integrating perception, planning, and control within a Dockerized ROS2 environment, this project demonstrates practical applications of robotics software and highlights the potential of autonomous systems in real-world, high-risk scenarios.

Project Video

Docker Instructions

  1. Build Docker Image:
docker build -f <DockerfileName> -t <YourDockerImage:Tag> .
  1. Create & run docker container from image:
xhost +local:root &&
docker run -it --rm \
            -e DISPLAY=$DISPLAY \
            -e QT_X11_NO_MITSHM=1 \
            -v /tmp/.X11-unix:/tmp/.X11-unix:rw \
            --device=/dev/dri \
            --privileged \
            <YourDockerImage:Tag>
  1. open container in different window:
docker exec -it <ContainerID> bash

#Note: Find the ID or name using: docker ps

Working Inside the Docker Container

Linorobot2 Simple Instructions

  1. Start Gazebo with 4WD base in Building.world: ros2 launch linorobot2_gazebo gazebo.launch.py robot_base:=4wd use_sim_time:=true world:=/ros2_ws/src/worlds/Building.world

  2. Start RViz with robot description and TF: ros2 launch linorobot2_description description.launch.py rviz:=true use_sim_time:=true

    Optional(Gazebo does this): Launch EKF for Odometry Fusion: ros2 launch linorobot2_navigation ekf.launch.py use_sim_time:=true

SLAM (Map Creation)

  1. Start SLAM Toolbox to build a map: ros2 launch slam_toolbox online_async_launch.py

  2. Drive the robot manually using teleop (in a separate terminal): ros2 run teleop_twist_keyboard teleop_twist_keyboard Note: Explore the map fully before saving

  3. Save the generated map: ros2 run nav2_map_server map_saver_cli -f /ros2_ws/src/maps/map

Autonomous Navigation (Using Saved Map)

  1. TEMPORARILY publish a static transform from map to odom (required to use 'map' as the fixed frame in RViz): ros2 run tf2_ros static_transform_publisher 0 0 0 0 0 0 map odom

  2. Launch the full Nav2 stack with your saved map: ros2 launch nav2_bringup bringup_launch.py map:=/ros2_ws/src/maps/map.yaml use_sim_time:=true

  3. In RViz:

    • Use the "2D Pose Estimate" tool to set the robot's rough location on the map.
    • Make sure the "Fixed Frame" is set to "map".

  4. In RViz:

    • Use the "Nav2 Goal" (2D Nav Goal) tool to send a destination on the map.
    • Watch the robot navigate autonomously!

Notes:

  • Run each launch command in a separate terminal inside the container.
  • Ensure Gazebo and RViz have access to your display (e.g., use -e DISPLAY=$DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix when running the container).
  • If AMCL reports pose errors, ensure the initial pose is in the "map" frame and simulation time is being used.